add functionality for apple data

app.py

@@ -5,14 +5,24 @@ import numpy as np
 import gradio as gr
 from models.inception import *
 from scipy.signal import resample
+import pandas as pd
 
 
+
+def load_apple_data():
+    return None
+
+
+
+# load wfdb data
 def load_data(sample_data):
     ecg, meta_data = wfdb.rdsamp(sample_data)
     lead_I = ecg[:,0]
     sample_frequency = meta_data["fs"]
     return lead_I, sample_frequency
 
+
+
 def preprocess_ecg(ecg,fs):
     if fs != 100:
         ecg = resample(ecg, int(len(ecg)*(100/fs)))
@@ -43,13 +53,23 @@ def run(header_file, data_file):
     SAMPLE_FREQUENCY = 100
     TIME = 10
     NUM_LEADS = 1
+    NEW_SAMPLE_FREQUENCY = 100
     demo_dir = f"{CWD}/sample_data"
-    _, hdr_basename = os.path.split(header_file.name)
-    _, data_basename = os.path.split(data_file.name)
-    shutil.copyfile(data_file.name, f"{demo_dir}/{data_basename}")
-    shutil.copyfile(header_file.name, f"{demo_dir}/{hdr_basename}")
-    data, fs = load_data(f"{demo_dir}/{hdr_basename.split('.')[0]}")
-    ecg = preprocess_ecg(data,fs)
+
+    if data_file.name.endswith(".csv"):
+        _, data_basename = os.path.split(data_file.name)
+        shutil.copyfile(data_file.name, f"{demo_dir}/{data_basename}")
+        df= pd.read_csv(f"{demo_dir}/{data_basename}", skiprows=12, sep=";", header=None,  decimal=',')
+        ecg = np.asarray(df[0].str.replace(',', '.').str.replace('−', '-').astype(float))
+        ecg = resample(ecg[(TIME*SAMPLE_FREQUENCY):(TIME*SAMPLE_FREQUENCY*2)],TIME * NEW_SAMPLE_FREQUENCY)
+        ecg = ecg/1000
+    else:
+        _, hdr_basename = os.path.split(header_file.name)
+        _, data_basename = os.path.split(data_file.name)
+        shutil.copyfile(data_file.name, f"{demo_dir}/{data_basename}")
+        shutil.copyfile(header_file.name, f"{demo_dir}/{hdr_basename}")
+        data, fs = load_data(f"{demo_dir}/{hdr_basename.split('.')[0]}")
+        ecg = preprocess_ecg(data,fs)
     age_model = load_age_model(sample_frequency=SAMPLE_FREQUENCY,recording_time=TIME,num_leads=NUM_LEADS)
     gender_model = load_gender_model(sample_frequency=SAMPLE_FREQUENCY,recording_time=TIME,num_leads=NUM_LEADS)
     age_estimate = age_model.predict(np.expand_dims(ecg,0)).ravel()[0]
@@ -66,7 +86,7 @@ with gr.Blocks() as demo:
     with gr.Row():
         with gr.Column(scale=1):
             header_file = gr.File(label = "header_file", file_types=[".hea"],)
-            data_file = gr.File(label = "data_file", file_types=[".dat"])
+            data_file = gr.File(label = "data_file", file_types=[".dat",".csv"])
         with gr.Column(scale=1):
             output_age = gr.Textbox(label = "Estimated age")
             output_gender = gr.Label( label = "Predicted gender")

apple_test.py

@@ -0,0 +1,41 @@
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+from scipy import signal
+import os
+import wfdb
+import shutil
+import gradio as gr
+from models.inception import *
+
+SAMPLE_FREQUENCY = 512
+TIME = 10
+NEW_SAMPLE_FREQUENCY = 100
+
+def load_age_model(sample_frequency,recording_time, num_leads):
+    cwd = os.getcwd()
+    weights = f"{cwd}/models/weights/model_weights_leadI_age.h5"
+    model = build_age_model((sample_frequency * recording_time, num_leads), 1)
+    model.load_weights(weights)
+    return model
+
+def load_gender_model(sample_frequency,recording_time, num_leads):
+    cwd = os.getcwd()
+    weights = f"{cwd}/models/weights/model_weights_leadI_gender.h5"
+    model = build_gender_model((sample_frequency * recording_time, num_leads), 1)
+    model.load_weights(weights)
+    return model
+
+age_model = load_age_model(100,10,1)
+gender_model = load_gender_model(100,10,1)
+ecg_path = "../../apple_health_export/electrocardiograms/"
+for i in os.listdir(ecg_path):
+    df= pd.read_csv(os.path.join(ecg_path,i), skiprows=12, sep=";", header=None,  decimal=',')
+    ecg = np.asarray(df[0].str.replace(',', '.').str.replace('−', '-').astype(float))
+    new_ecg = signal.resample(ecg[(TIME*SAMPLE_FREQUENCY):(TIME*SAMPLE_FREQUENCY*2)],TIME * NEW_SAMPLE_FREQUENCY)
+    # from uV to mV
+    new_ecg = new_ecg/1000
+    age_estimate = age_model.predict(np.expand_dims(new_ecg,0)).ravel()[0]
+    gender_prediction = gender_model.predict(np.expand_dims(new_ecg,0)).ravel()[0]
+    print("Estimated age: ", age_estimate)
+    print("Predicted gender: Male = {}, Female = {}".format((1- gender_prediction),(gender_prediction)))